28.11.2014 Views

May-June 2013 Issue - American College of Phlebology

May-June 2013 Issue - American College of Phlebology

May-June 2013 Issue - American College of Phlebology

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

MAY-JUNE <strong>2013</strong><br />

Deactivation<br />

<strong>of</strong> sodium<br />

tetradecyl<br />

sulphate<br />

injection by<br />

blood proteins<br />

PAGE 8<br />

Anatomy and<br />

embryology <strong>of</strong> the<br />

small saphenous<br />

vein, nerve<br />

relationships and<br />

implications for<br />

treatment<br />

PAGE 14


Deactivation <strong>of</strong> sodium<br />

tetradecyl sulphate<br />

injection by blood<br />

proteins<br />

contents<br />

may-jun ‘13<br />

Contributing Editor/Reviewer:<br />

From the<br />

Editor-in-Chief<br />

Dr. Nick Morrison 5<br />

Neil Sadick, MD, FACP, FACPh, FAAD, FAACS<br />

Associate Editor:<br />

Ted King MD, FAAFP, FACPh<br />

8<br />

Foot-sparing<br />

postoperative<br />

compression bandage:<br />

a possible alternative<br />

to the traditional<br />

bandage<br />

Anatomy and<br />

embryology <strong>of</strong> the<br />

small saphenous vein,<br />

nerve relationships<br />

and implications for<br />

treatment<br />

Contributing Editor/Reviewer:<br />

Hugo Partsch, MD<br />

Associate Editor:<br />

Mitchell Goldman, MD, FACPh<br />

Contributing Editor/Reviewer:<br />

Mark Isaacs, MD<br />

Associate Editor:<br />

11 14<br />

Pauline Raymond-Martimbeau, MD, FACPh<br />

Prevention <strong>of</strong> venous<br />

thromboembolism:<br />

the Seventh ACCP<br />

Conference on<br />

Antithrombotic and<br />

Thrombolytic Therapy<br />

Contributing Editor/Reviewer:<br />

Amjad T. AlMahameed, MD, MPH<br />

Associate Editor:<br />

Stephanie Dentoni, MD<br />

17


the world<br />

<strong>of</strong> vein care<br />

unites in Boston this September<br />

Hosted by the <strong>American</strong> <strong>College</strong> <strong>of</strong> <strong>Phlebology</strong> in conjunction with the International Union <strong>of</strong> <strong>Phlebology</strong>, the XVII UIP<br />

World Meeting will bring together respected faculty, physicians and health care pr<strong>of</strong>essionals from across the globe<br />

this September 8-13 to showcase the most advanced research, technology and treatments in the field <strong>of</strong> vein care.<br />

This truly historic event will include sessions covering the wide<br />

spectrum <strong>of</strong> phlebology, including:<br />

+ More than 300 internationally recognized speakers<br />

from around the world<br />

+ Innovative and clinically relevant educational sessions<br />

+ Hands-on simulation workshops and demonstrations<br />

with international faculty<br />

+ More than 85 event exhibitors showcasing the world's<br />

most advanced products, pharmaceuticals and<br />

medical devices<br />

+ Opportunities for all levels <strong>of</strong> skill, from basic through<br />

advanced, to improve your practice and patients' care<br />

Join the ACP at the XVII UIP World Meeting in Boston, September 8-13 for this extraordinary event.<br />

UNION INTERNATIONALE DE PHLEBOLOGIE<br />

INTERNATIONAL UNION OF PHLEBOLOGY<br />

World Meeting <strong>of</strong> the International Union <strong>of</strong> <strong>Phlebology</strong><br />

/// September 8–13, <strong>2013</strong><br />

Hynes Convention Center • Boston, Massachusetts • USA<br />

HOSTED BY<br />

510.346.6800 | www.uip<strong>2013</strong>.org | www.phlebology.org<br />

advancing vein care


<strong>2013</strong> Compression Hosiery and<br />

support Wear Buyer’s Guide<br />

n Everything you need to know about over-the-counter<br />

and medical compression therapy products<br />

n Photos and detailed product information from eleven<br />

manufacturers<br />

n Compression stockings, pantyhose, thigh-highs,<br />

maternity, travel and sports socks, plus lymphedema<br />

products and armsleeves<br />

n Direct-from-the-factory discounts for individuals or<br />

medical pr<strong>of</strong>essionals<br />

www.BrightLifeDirect.com<br />

Your one-stop shop for<br />

compression therapy.<br />

to request your FREE copY<br />

or Call 1-877-545-8585<br />

6925-D Willow Street NW • Washington, DC 20012<br />

Find tREatmEnt solutions FoR<br />

Tired & Achy Legs • Varicose Veins • Vascular Disease • Swelling & Lymphedema • Ulcers • Fragile/Sensitive Skin • DVT Prevention<br />

View past issues <strong>of</strong> <strong>Phlebology</strong> Forum at<br />

www.phlebology.org


disclosure<br />

<strong>of</strong> interests<br />

Name ACP Role Date<br />

Submitted<br />

Disclosure<br />

Stephanie Dentoni, MD Recruitment & Retention (Chair) 6/25/12 Nothing to Disclose<br />

Mark Forrestal, MD, FACPh ACP 6/25/12 New Star Lasers Cooltouch: Speaker, Trainer<br />

Mitchel Goldman, MD,<br />

FACPh<br />

<strong>Phlebology</strong> Forum Task Force 2/14/<strong>2013</strong> <strong>American</strong> Society for Dermatologic Surgery,<br />

President-Elect; Merz Aesthetics/Kruesler,<br />

Consultant<br />

Jean-Jerome Guex, MD,<br />

FACPh<br />

ACP BOD, Communications,<br />

Standing Committee,<br />

Leadership Development, UIP<br />

<strong>2013</strong> Task Force, AMA HOD<br />

Task Force, International Affairs<br />

(Chair),<br />

6/14/12 Kreussler: Speaker; Sigvaris: Speaker, Investigator,<br />

Consultant; Innotech: Principal Investigator; Pierre<br />

Fabre: Consultant; Boerighr Ingelheim: Consultant,<br />

Medical Writer; Servier: Investigator, Consultant,<br />

Speaker<br />

Lowell Kabnick, MD, FACS,<br />

FACPh<br />

UIP <strong>2013</strong> Task Force 7/17/12 Angiodynamics: Consultant, Shareholder, Patent;<br />

Vascular Insights: Scientific Advisory Board<br />

Neil Khilnani, MD, FACPh<br />

ACP BOD, Member Services<br />

(Chair)<br />

7/24/12 Sapheon: Data Safety Board Member<br />

Ted King, MD, FAAFP, FACPh<br />

ACP BOD, Leadership<br />

Development, PES-QM Task<br />

Force, Public Education<br />

6/14/12 BTG: Investigator; Merz: Speaker<br />

Mark Meissner, MD<br />

ACP BOD, Education Standing<br />

Committee<br />

7/13/12 Nothing to Disclose<br />

Nick Morrison, MD, FACS,<br />

FACPh<br />

UIP <strong>2013</strong> Task Force (Chair),<br />

<strong>Phlebology</strong> Forum Task Force<br />

(Chair), Annual Congress<br />

Planning Committee (Chair)<br />

6/13/12 Medi: Speakers Bureau; Merz: Speakers Bureau;<br />

Sapheon: Principle Investigator; VeinX: Scientific<br />

Advisory Board<br />

Eric Mowatt-Larssen, MD ACP CME Committee 6/25/12 BTG International, Inc.: Consultant<br />

Diana Neuhardt, RVT, RPhS<br />

ACP BOD, Member Services,<br />

Audit, UIP <strong>2013</strong> Task Force,<br />

<strong>Phlebology</strong> Forum Task<br />

Force, Veinline, Recruitment<br />

& Tetention, CME, Distance<br />

Learning, Public Education<br />

(Chair)<br />

6/15/12 Nothing to Disclose<br />

Pauline Raymond-<br />

Martimbeau, MD, FACPh<br />

UIP <strong>2013</strong> Task Force 6/22/12 Nothing to Disclose<br />

6


From the<br />

Editor-in-Chief<br />

Dear Readers<br />

In this issue <strong>of</strong> <strong>Phlebology</strong> Forum you will find a number <strong>of</strong> topics<br />

related to deep and superficial venous problems, many <strong>of</strong> which expand<br />

our understanding <strong>of</strong> what we see and can do on a practical basis in<br />

daily work. Included are an investigation into the deactivation <strong>of</strong> STS<br />

sclerosant which may help explain the varied clinical outcomes we all see;<br />

an introduction to the vagaries <strong>of</strong> small saphenous vein and associated<br />

nerve anatomy and their clinical implication; relevant clinical information<br />

regarding post-treatment compression therapy which adds to the dogmachallenging<br />

discussion <strong>of</strong> compression; and an excellent analysis <strong>of</strong> the<br />

ACCP guidelines for thromboprophylaxis.<br />

And finally, I will beat the drum on behalf <strong>of</strong> the ACP and the Organizing<br />

Committees for the UIP XVII World Congress in Boston, and I can promise<br />

this meeting will be one <strong>of</strong> the best “vein” meetings you have ever<br />

attended and will greatly exceed your expectations. The countdown is<br />

now less than 3 months and I have only three other things to say about<br />

the meeting: BOSTON! BOSTON! BOSTON!<br />

Nick Morrison, MD<br />

Editor-in-Chief<br />

<strong>Phlebology</strong> Forum<br />

7


Deactivation <strong>of</strong> sodium<br />

tetradecyl sulphate<br />

injection by blood proteins<br />

Watkins MR.<br />

Eur J Vasc Endovasc Surg. 2011 Apr;41(4):521-5. doi: 10.1016/j.ejvs.2010.12.012. Epub 2011 Jan 22.<br />

Contributing Editor/Reviewer: Neil Sadick, MD, FACP, FACPh, FAAD, FAACS<br />

Associate Editor: Ted King MD, FAAFP, FACPh


COMMENTARY<br />

The clinical phlebologist is well aware that<br />

sclerosants act at limited areas <strong>of</strong> clinical activity<br />

from the point where they are injected. In the past,<br />

it was felt that this clinical observation was related<br />

to the dilution <strong>of</strong> the sclerosant and its subsequent<br />

decreased potency in this regard, based solely upon<br />

concentration/sclerosant endothelial interaction<br />

kinetically.<br />

in practical use, it<br />

would take less than<br />

0.5% ml <strong>of</strong> whole<br />

blood to deactivate<br />

1 ml <strong>of</strong> 3% STS<br />

In addition, a commonly known observation is that foam sclerosants manifest greater potency and also have more significant<br />

distal watershed-type effect. With these tenets in mind, the article entitled “Deactivation <strong>of</strong> sodium tetradecyl sulphate<br />

injection by blood protein” brings new insight and rationale into the actual pathophysiologic mechanisms <strong>of</strong> this clinical<br />

observation.<br />

The present study looked at the volume <strong>of</strong> blood required to inactivate 1ml <strong>of</strong> 3% sodium tetradecyl sulphate. This was<br />

accomplished by measuring the concentration <strong>of</strong> sodium tetradecyl sulphate (STS) remaining in the active state in an in-vitro<br />

stock solution after adding increasing volumes <strong>of</strong> solution containing either bovine serum albumin or red blood cells or a<br />

mixture <strong>of</strong> both components. The two methodologies utilized to assess the results included autotritation and colorimetry.<br />

Results <strong>of</strong> the aforementioned study showed that STS is deactivated by blood proteins in a linear fashion. The protein<br />

solution utilized seemed to make little difference, with approximately 2 ml <strong>of</strong> 4% protein solution deactivating 1 ml <strong>of</strong> a 3%<br />

STS solution.<br />

The authors hypothesized that titration measured the free STS remaining in solution after the addition <strong>of</strong> blood proteins<br />

which would bind and deactivate the STS and this correlates with clinical activity. These results suggest that in practical<br />

use, it would take less than 0.5% ml <strong>of</strong> whole blood to deactivate 1 ml <strong>of</strong> 3% STS which would explain why an empty vein<br />

technique is important when injecting liquid sclerosant in the varicose veins, a concept that is well-documented in clinical<br />

practice since its description by Orbach. 1<br />

This would explain the noted limited distance <strong>of</strong> clinical efficacy <strong>of</strong> injected liquid sclerosants.<br />

Limitation <strong>of</strong> this study would include its in-vitro design and qualitative visual assessment as related to the colorimetric<br />

assay employed in the study design. The authors also confirm that it was difficult to visually determine the end point <strong>of</strong><br />

titration with increasing volumes <strong>of</strong> blood proteins, which is why the average <strong>of</strong> up to four individual readings was utilized.<br />

1 Orbach EJ. Sclerotherapy <strong>of</strong> varicose veins—utilization <strong>of</strong> an intravenous air block. Am J Surg 1944;LXVI(3):362–6.<br />

9


Therefore, the results should be interpreted as a guide rather than an exact figure.<br />

The aforementioned article is <strong>of</strong> great clinical importance to the practicing phlebologist although it is not without<br />

limitations in terms <strong>of</strong> its design. This is one <strong>of</strong> the few articles in the literature which would attempt to understand<br />

physiologic interactions between liquid sclerosants and blood vessels and helps to explain many <strong>of</strong> the clinical observations<br />

noted by the practicing phlebologist.<br />

The characteristics <strong>of</strong> sclerosants that the practicing phlebologist considers in choosing an agent for daily clinical practice<br />

include efficacy, which is <strong>of</strong>ten dependent upon its strength and thus the ability to induce pan endothelial destruction and<br />

its distal downstream clinical efficacy from the site <strong>of</strong> initial injection. 2 Other considerations include its minimal sclerosant<br />

concentration, which will achieve pan endothelial destruction. 3 Allergenicity and side effect pr<strong>of</strong>iles are important<br />

considerations in this regard. 4<br />

Other investigators have utilized cell lysis and human blood plasma looking at the lytic effect <strong>of</strong> STS on red blood cells,<br />

platelets and cultured endothelial cells. 5 The main difference between the other studies and this work is that in other studies<br />

the lytic effects <strong>of</strong> the sclerosants on blood and endothelial cells in the presence <strong>of</strong> blood proteins were measured, whereas<br />

the present study quantifies the residual STS remaining after the addition <strong>of</strong> protein in a standard STS solution. However,<br />

with analysis <strong>of</strong> both sets <strong>of</strong> results it is clear that a low volume <strong>of</strong> blood (0.50 ml) is enough to neutralize 1 ml <strong>of</strong> 3% STS.<br />

There are several clinical implications that can be taken from these studies. First, displacement <strong>of</strong> blood prior to treating a<br />

vein (Empty Vein Technique) may improve clinical efficacy. Second, foam sclerotherapy, which displaces more blood than<br />

liquid sclerosants, may explain its greater efficacy. However, this is less obvious with larger vessels where more treatment<br />

may be necessary because foam is less efficient at displacing blood as veins get larger.<br />

Finally, repeated puncture instillations at frequent intervals may be more effective in treating larger veins because <strong>of</strong> the<br />

displacement <strong>of</strong> blood, and the practitioner may even consider utilizing tumescent techniques to decrease red blood cell<br />

volume as an additional augmenting maneuver. 6<br />

In summary, it appears from this and other published studies and clinical experience that it is a combination <strong>of</strong> chemical<br />

interactions and host factors that determine clinical outcomes with regard to the utilization <strong>of</strong> liquid sclerosants.<br />

2 Murad MH, Coto-Yglesias F, Zumaeta-Garcia M, Elamin MB, Duggirala MK, Erwin PJ, Montouri VM, Gloviczki P. A Systematic review and meta-analysis <strong>of</strong><br />

the treatments <strong>of</strong> varicose veins. J. Vasc Surg. 2011; 53(5 Suppl): 495-655.<br />

3 Sadick NS, Choosing the appropriate sclerosant concentration for vessel diameter. Dermatol Surg 2010;36 Suppl 2976-981.<br />

4 Cavezzi A, Parsi K. Complications <strong>of</strong> foam sclerotherapy. <strong>Phlebology</strong> 2012; 27 Suppl 146-51.<br />

5 Parsi K, Exner R, Conner DE, Herbert A, Ma DDF, Joseph JE. The lytic effects <strong>of</strong> detergent sclerosants on erythrocytes, platelets, endothelial cells and<br />

microparticles are attenuated by albumin and other plasma components in vitro. Eur J Vasc Endovasc Surg 2008; 36(2): 216-223.<br />

6 Ramelet A-A. Sclerotherapy in tumescent anesthesia <strong>of</strong> reticular veins and telangiectasia’s. Dermatol Surg 2012; 38(5): 748-751.<br />

10


Foot-sparing postoperative<br />

compression bandage:<br />

a possible alternative to the<br />

traditional bandage<br />

Author: Ricci S, Moro L, Trillo L, Incalzi RA.<br />

<strong>Phlebology</strong>. <strong>2013</strong> Feb;28(1):47-50.<br />

Contributing Editor/Reviewer: Hugo Partsch, MD<br />

Associate Editor: Mitchell Goldman, MD, FACPh<br />

11


Abstract<br />

Objectives: the aim <strong>of</strong> this study was to evaluate the<br />

efficacy and tolerability <strong>of</strong> foot sparing bandages after<br />

varicose vein surgery performed in CEAP C2 patients.<br />

Methods: 90 consecutive lower extremities in 129<br />

patients, for whom different kinds <strong>of</strong> varicose vein<br />

surgery were performed, received inelastic footsparing<br />

bandages for one week. The bandages consist<br />

<strong>of</strong> thin polyurethane under- wraps, cotton pads<br />

selectively placed over operated tracks and <strong>of</strong> 10 cm<br />

wide adhesive bandages (Fortelast® Lohmann). In order<br />

to prevent slippage acrylic glue wraps are attached<br />

to the skin at the proximal end <strong>of</strong> the bandages.<br />

Patient’s satisfaction, efficacy and local effects were<br />

systematically documented.<br />

Results: High satisfaction was achieved. Measurement<br />

<strong>of</strong> the bandage pressures showed mean values <strong>of</strong> 25<br />

foot-sparing<br />

inelastic bandages<br />

after varicose<br />

vein surgery are<br />

effective, cheap<br />

and well tolerated<br />

at least in a large<br />

proportion <strong>of</strong><br />

mobile and active<br />

patients<br />

mmHg are reported in the supine position, 30 mmHg<br />

in standing and 35 mmHg during walking. Four <strong>of</strong> the<br />

first 20 cases experienced a slight morning oedema <strong>of</strong> the foot, which disappeared while walking. Therefore, in the remaining<br />

cases the foot and distal limb were covered with a custom short tubular-shaped ‘sock’ providing 10 mmHg compression, only<br />

during the first 24 hours.<br />

Conclusion: The described foot-sparing inelastic bandages after varicose vein surgery are effective, cheap and well tolerated<br />

at least in a large proportion <strong>of</strong> mobile and active patients who do not show skin changes in the gaiter area.<br />

Commentary<br />

In the last few years the value <strong>of</strong> compression after varicose vein procedures has been questioned by several investigators<br />

who used compression hosiery. Even TED stockings were applied which are definitely too weak to reduce the venous diameter<br />

<strong>of</strong> disconnected superficial veins in the upright position. 1 On the other hand strongly applied conventional bandages<br />

1 Partsch H, Mosti G, Uhl JF. Unexpected venous diameter reduction by compression stocking <strong>of</strong> deep, but not <strong>of</strong> superficial veins. Veins and Lymphatics<br />

2012; 1:e3 online.<br />

12


in addition to the application <strong>of</strong> local pads over the treated areas were shown to have considerable benefits concerning a<br />

reduction <strong>of</strong> pain and hematoma formation. 2 , 3<br />

The reported method <strong>of</strong> applying bandages starting above the ankle and sparing the foot <strong>of</strong>fers especially the advantages<br />

that the movement <strong>of</strong> the ankle joint remains unimpeded and that patients are able to put on and to wear their normal<br />

shoes. Slight edema distal to the bandage may occur after rest but disappears quickly as soon as the patient starts walking.<br />

As the authors have shown this swelling may be prevented by light compression socks.<br />

Using inelastic compression material <strong>of</strong>fers the advantage <strong>of</strong> achieving higher pressures in the upright position and during<br />

walking while the resting pressure stays in a comfortable range.<br />

We use foot-sparing compression bandages mainly after (foam-) sclerotherapy <strong>of</strong> large tributaries and after phlebectomy in<br />

the same way as the authors, just with higher pressures.<br />

The positive experience <strong>of</strong> the authors challenge the dogma that leg compression needs to provide a pressure gradient and<br />

shows that in mobile patients, even without any compression <strong>of</strong> the gaiter area, no adverse “strangulation effects” occur.<br />

It has been demonstrated that compression pressure which is higher at calf than at gaiter level is also more effective in<br />

enhancing venous pump function 4 .<br />

We are grateful to the authors for reminding us <strong>of</strong> this very efficient way <strong>of</strong> compression which can be used in a large<br />

proportion <strong>of</strong> patients after venous stripping, phlebectomy and endovenous procedures.<br />

2 Lugli M, Cogo A, Guerzoni S, Petti A, Maleti O. Effects <strong>of</strong> eccentric<br />

compression by a crossed-tape technique after endovenous laser ablation <strong>of</strong> the great saphenous vein: a randomized study. <strong>Phlebology</strong>. 2009<br />

Aug;24(4):151-6.<br />

3 Mosti G, Mattaliano V, Arleo S, Partsch H. Thigh compression after great saphenous surgery is more effective with high pressure. Int Angiol. 2009<br />

Aug;28(4):274-80.<br />

4 Mosti G, Partsch H. High compression pressure over the calf is more effective than graduated compression in enhancing venous pump function. Eur J Vasc<br />

Endovasc Surg. 2012 Sep;44(3):332-6.<br />

13


Anatomy and embryology <strong>of</strong> the small<br />

saphenous vein, nerve relationships<br />

and implications for treatment<br />

Author: Uhl JF, Gillot C<br />

<strong>Phlebology</strong> <strong>2013</strong>;28:4-15<br />

Contributing Editor/Reviewer: Mark Isaacs, MD<br />

Associate Editor: Pauline Raymond-Martimbeau, MD, FACPh<br />

14


COMMENTARY<br />

Treatment <strong>of</strong> the small saphenous vein is known to be challenging, both because <strong>of</strong> the highly variable anatomy <strong>of</strong><br />

the vein and because <strong>of</strong> vulnerability <strong>of</strong> adjacent nerves and arteries to injury. Sural nerve injury and post operative<br />

paresthesia are two <strong>of</strong> the most common complications <strong>of</strong> treatment. 1 , 2 Surgeons traditionally have limited excision<br />

<strong>of</strong> the vein to a proximal segment and rarely if ever attempt to fully expose the popliteal vein in an effort to prevent<br />

neurological or arterial complications. This anatomical<br />

review by Uhl and Gillot is fascinating in its detail<br />

and highly relevant to the surgeon or nonsurgeon<br />

phlebologist approaching treatment <strong>of</strong> the small<br />

saphenous vein. The authors are to be commended for<br />

providing such an enlightening treatise on the known<br />

embryology and anatomy <strong>of</strong> the vein, with clearly<br />

delineated correlations between the described anatomy<br />

and possible treatment complications.<br />

Going beyond the usual text descriptions, the<br />

authors make liberal use <strong>of</strong> diagrams and well labeled<br />

anatomical dissections to illustrate “high risk zones”:<br />

the ankle, the “apex <strong>of</strong> the calf” 3 , and the popliteal<br />

fossa. For example, at its termination the arch <strong>of</strong> the<br />

small saphenous vein crosses close to the tibial and<br />

medial gastrocnemius nerves. The sural nerve and two<br />

“companion” vessels are shown to be in close proximity<br />

...a thorough<br />

knowledge <strong>of</strong> the<br />

small saphenous vein<br />

and it’s surrounding<br />

structures remains<br />

mandatory. This<br />

paper should have<br />

a prominent place<br />

in the library <strong>of</strong> all<br />

physicians treating<br />

small saphenous<br />

venous insufficiency<br />

to the small saphenous vein below the calf muscles. In<br />

the ankle the origin <strong>of</strong> the vein is deep below the fascia<br />

and plexiform, with the nerve once again very closely associated.<br />

The authors, however, go beyond their anatomical descriptions to draw conclusions about clinical practice. They state,<br />

for example, that the possible existence <strong>of</strong> a short saphenous artery poses a “high risk for injection <strong>of</strong> a sclerosing<br />

1 Van Groenendael L, et al. Conventional surgery and endovenous laser ablation <strong>of</strong> recurrent varicose veins <strong>of</strong> the small saphenous vein: a retrospective<br />

clinical comparison <strong>of</strong> assessment <strong>of</strong> patient satisfaction. Plebology 2010;25:151-157<br />

2 Gibson K, et al. Endovenous laser treatment <strong>of</strong> the short saphenous vein: Efficacy and complications. J Vasc Surg April 2007 45;4:795-800<br />

3 In order to prevent confusion amongst English language readers it should be noted that “the apex <strong>of</strong> the calf” is an anatomical term used in several<br />

places in this paper by Gillot to refer to the location <strong>of</strong> the junction between the muscle <strong>of</strong> the calf and the Achille’s tendon, not the highest point <strong>of</strong><br />

the calf.<br />

15


agent”, that “a good rule for endovenous laser treatment is to remain at least 4 cm below the popliteal crease to<br />

avoid a thermal injury <strong>of</strong> the nerves” and that a detailed “triple check” by duplex ultrasound <strong>of</strong> venous flow and<br />

nearby nerves and arteries is the “best way to reduce the risk <strong>of</strong> severe arterial or nerve complications.” While such<br />

conjecture is very reasonable and intuitively appealing given the anatomical relationships so carefully delineated in the<br />

article, there is a lack <strong>of</strong> documentation to support these conclusions. Regarding the recommendation for the type <strong>of</strong><br />

duplex examination they describe, the scarcity <strong>of</strong> ultrasonographers adequately trained and skilled enough to identify<br />

not only superficial veins and tributaries but nearby nerves and small arteries makes following their recommendations<br />

utopian. As one author put it, nerves in the area <strong>of</strong> the popliteal fossa are, “always visible, never seen”. 4<br />

While good comparative studies are lacking, a recent review <strong>of</strong> the literature on treatment <strong>of</strong> the small saphenous<br />

vein would appear to confirm that endovenous treatment techniques including radi<strong>of</strong>requency heat ablation, laser<br />

ablation and ultrasound guided foam sclerotherapy carry less risk <strong>of</strong> complications than surgery with better treatment<br />

outcomes. 5 One prominent surgeon has gone so far as to declare surgery for small saphenous reflux to be obsolete! 6<br />

The explanation may lie not only with the lack <strong>of</strong> perivenous trauma but also with the use <strong>of</strong> tumescent anesthesia.<br />

We know that the advent <strong>of</strong> the use <strong>of</strong> tumescent anesthesia prior to endovenous catheter heat ablation treatment<br />

<strong>of</strong> the great saphenous vein resulted in a drop in neurological complications, presumably because properly injected<br />

tumescent fluid not only provides anesthesia, vein compression and a fluid heat “sink” but possibly also because it<br />

peels adjacent nerves and arteries away from the wall <strong>of</strong> the target vein. One might legitimately question whether<br />

the authors’ clinical guidelines are completely justified when tumescent anesthesia is properly administered prior to<br />

treatment. Certainly it is better to err on the side <strong>of</strong> caution, however, by keeping in mind the authors’ high risk zones<br />

even when liberal tumescent anesthesia has been used.<br />

Unfortunately, there are many surgeons who have adopted endovenous treatment techniques but are unwilling or<br />

unable to wean themselves from the operating room and general anesthesia. Likewise, proper administration <strong>of</strong><br />

tumescent anesthesia under ultrasound guidance is a challenging skill to learn even for those motivated to adopt it,<br />

and heat ablation done under tumescent anesthesia is far from risk free. Some practitioners do not use direct real-time<br />

ultrasound guidance during their procedures. For these reasons amongst others a thorough knowledge <strong>of</strong> the small<br />

saphenous vein and its surrounding structures remains mandatory. This paper should have a prominent place in the<br />

library <strong>of</strong> all physicians treating small saphenous vein insufficiency.<br />

4 Ricci S. Ultrasound Observation <strong>of</strong> the Sciatic Nerve and Its Branches at the Popliteal Fossa: Always Visible, Never Seen. Eur J Vasc Endovasc Surg<br />

2005;30:659-663<br />

5 Tellings SS, Ceulen PPM, Sommer A. Surgery and endovenous techniques for the treatment <strong>of</strong> small saphenous varicose veins: a review <strong>of</strong> the literature,<br />

<strong>Phlebology</strong> 2011;26:179-184<br />

6 Myers K. Surgery for small saphenous reflux is obsolete! Venous Digest 2005, modified from the ANZ J <strong>Phlebology</strong><br />

16


Prevention <strong>of</strong> venous<br />

thromboembolism:<br />

the Seventh ACCP Conference<br />

on Antithrombotic and<br />

Thrombolytic Therapy<br />

Author: Geerts WH, Bergqvist D, Pineo GF, Heit JA, et al.<br />

Chest. 2004;126(suppl):338S-400S.<br />

Contributing Editor/Reviewer: Amjad T. AlMahameed, MD, MPH<br />

Associate Editor: Stephanie Dentoni, MD<br />

17


COMMENTARY<br />

The 9th <strong>American</strong> <strong>College</strong> <strong>of</strong> Chest Physicians (ACCP)<br />

guidelines for prevention <strong>of</strong> venous thromboembolic<br />

disease (VTE) noted a major shift in the methodology<br />

used by the authors. They discarded surrogate endpoints<br />

(asymptomatic venographically-evident VTE) and elevated<br />

death and symptomatic VTE as the only reliable outcome<br />

measures. They also advocate incorporating bleeding risk<br />

alongside thrombotic risk assessment when deciding on<br />

pharmacoprophylaxis. Lastly, they recommend against the<br />

application <strong>of</strong> performance measures in medical patients<br />

that promote universal VTE prophylaxis regardless <strong>of</strong> risk.<br />

In the present viewpoint, I also review the events that<br />

led to mandating performance measures that, despite<br />

best intensions, may have promoted unforeseen risks.<br />

While I disagree with discarding a large body <strong>of</strong> evidence<br />

that links asymptomatic VTE to clinically relevant major<br />

adverse events, including death, I agree with the final<br />

(VTE) remains...<br />

the most<br />

common cause<br />

<strong>of</strong> preventable<br />

hospital death,<br />

and a primary<br />

contributor<br />

to 10% <strong>of</strong> all<br />

hospital deaths.<br />

recommendations to use thromboprophylaxis based on<br />

individualized risk-benefit assessment.<br />

INTRODUCTION<br />

Venous thromboembolic disease (VTE) remains a major cause <strong>of</strong> morbidity and mortality, the most common cause <strong>of</strong><br />

preventable hospital death, and a primary contributor to 10% <strong>of</strong> all hospital deaths. 1 , 2 Overall, VTE may affect 10-40%<br />

<strong>of</strong> “at-risk” hospitalized patients who do not receive proper thromboprophylaxis. 1,3 In fact, the thrombotic risk extends<br />

well beyond discharge from the index hospitalization in both surgical and non-surgical patients. In one population-based<br />

retrospective study, the average annual age- and sex-adjusted incidence <strong>of</strong> in-hospital VTE was more than 100 times<br />

greater than the incidence among community residents. 4<br />

1 Geerts WH, Pineo GF, Heit JA, et al. Prevention <strong>of</strong> venous thromboembolism: the Seventh ACCP Conference on Antithrombotic and Thrombolytic<br />

Therapy. Chest. 2004;126(suppl):338S-400S.<br />

2 Sandler DA, Martin JF. Autopsy proven pulmonary embolism in hospital patients: are we detecting enough deep vein thrombosis? J R Soc Med.<br />

1989;82(4):203–205.<br />

3 Hillen HF. Thrombosis in cancer patients. Ann Oncol. 2000;11(suppl 3):273-276.<br />

4 Heit JA, Melton LJ 3rd, Lohse CM. Incidence <strong>of</strong> venous thromboembolism in hospitalized patients vs community residents. <strong>May</strong>o Clin Proc 2001;76:1102


How Did Thromboprophylaxis Become “Universal”?<br />

Several factors have led to adopting near universal thromboprophylaxis to hospitalized patients. These include the<br />

markedly increased VTE prevalence in this at-risk group, serious adverse consequences <strong>of</strong> VTE (death, pulmonary<br />

embolism [PE], chronic post-thrombotic pulmonary hypertension and post-phlebetic syndrome) and the proven efficacy,<br />

safety and cost-effectiveness <strong>of</strong> thromboprophylaxis. Other important factors that fostered the generalization <strong>of</strong> these<br />

universal thromboprophylaxis recommendations include: 1) Absence <strong>of</strong> discrete prodromal phase: most VTEs noted in<br />

thromboprophylactic trials are subclinical (asymptomatic). Even in patients that were eventually labeled as having had<br />

symptomatic VTE, their initial signs and symptoms are common in this population and non-specific. 2) Absence <strong>of</strong> reliable<br />

screening test: venography, the gold standard used in thromboprophylactic trials, is highly accurate. Nonetheless, it is an<br />

impractical, invasive, expensive, and potentially thrombogenic test. Venous duplex ultrasonography is expensive and not<br />

all that sensitive in a large portion <strong>of</strong> hospitalized patients.<br />

The past 10 years witnessed the introduction <strong>of</strong> a series <strong>of</strong> VTE prevention initiatives. Among these, several stand out<br />

as landmark documents: In 2006, The Joint Commission/National Quality Forum 17 performance measures (which was<br />

revised in 2008 to include 6 additional measures) were published. 5 , 6 In 2008, The ACCP evidence-based clinical practice<br />

guidelines on antithrombotic and thrombolytic therapy included an expanded section on VTE prevention. It strongly<br />

emphasized that hospitals should consider developing strategies to consistently identify hospitalized patients at risk for<br />

VTE and actively seek to prevent VTE occurrence and recurrence. 7 Soon thereafter, the US Surgeon General released a<br />

“Call to Action to Prevent Deep Vein Thrombosis and Pulmonary Embolism” that urged a coordinated, multifaceted plan<br />

to reduce the VTE incidence nationwide. 8 The Agency for Healthcare Research and Quality (AHRQ) contributed to the<br />

“Call to Action” with the release <strong>of</strong> two new guides: one for patients and another for health care providers, counseling<br />

both to participate in disseminating VTE prevention recommendations. 9<br />

These successive initiatives, along with well-publicized campaigns that followed, led to the rapid dissemination <strong>of</strong> these<br />

recommendations. Quickly, hospital administration became engaged and decided to support VTE prevention programs.<br />

As front-line clinicians embraced the recommendations, VTE prophylaxis practices became “standard <strong>of</strong> care” and<br />

5 http://www.qualityforum.org/Publications/2006/12/National_Voluntary_Consensus_Standards_for_Prevention_and_Care_<strong>of</strong>_Venous_<br />

Thromboembolism__Policy,_Preferred_Practices,_and_Initial_Performance_Measures.aspx<br />

6 http://www.qualityforum.org/Publications/2008/10/National_Voluntary_Consensus_Standards_for_Prevention_and_Care_<strong>of</strong>_Venous_<br />

Thromboembolism__Additional_Performance_Measures.aspx<br />

7 Geerts WH, Bergqvist D, Pineo GF, et al., <strong>American</strong> <strong>College</strong> <strong>of</strong> Chest Physicians. Prevention <strong>of</strong> venous thromboembolism: <strong>American</strong> <strong>College</strong> <strong>of</strong> Chest<br />

Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008; 133:381S–453S.<br />

8 http://www.surgeongeneral.gov/library/calls/deepvein/index.html<br />

9 http://www.ahrq.gov/pr<strong>of</strong>essionals/quality-patient-safety/patient-safety-resources/resources/vtguide/vtguide.pdf<br />

19


“clinical practice pathways”; while “computerized electronic alerts” as well as “algorithm-based” strategies were widely<br />

developed to ensure continued compliance.<br />

Another important, yet unspoken, factor that may have augmented this response was the federal government’s Centers for Medicare<br />

and Medicaid Services (CMS) threat to deny reimbursement for hospital-acquired VTE as it considered it a “preventable medical<br />

error”. Compliance with these guidelines (which the CMS presumed would eliminate such “preventable medical errors”) was therefore<br />

viewed as an important means towards improving patients’ outcomes, decreasing “preventable medical errors” and avoiding potential<br />

penalties. Such eagerness for “perfect compliance” was fostered and patrolled by hospital administrators and quality control medical<br />

<strong>of</strong>ficers. One wonders if this could have led to a speedy generalization <strong>of</strong> the recommendations to include some patients that may<br />

have been at extraordinarily low risk for VTE or unusually high risk for bleeding complications. Therefore, the “Sentinel Event Alert”<br />

released by The Joint Commission entitled “Preventing Errors Relating to Commonly Used Anticoagulants”) in late 2008 came as no<br />

surprise to us.<br />

Reconciling the Newly Adopted ACCP Outcome Measures with<br />

Daily Practice<br />

The ACCP methodologists attempted to weigh the evidence for and against thromboprophylaxis using prevention <strong>of</strong> fatal<br />

PE and “symptomatic” VTE (which they presumed to be synonymous with “clinically relevant” VTE) as the primary goal <strong>of</strong><br />

this treatment, while at the same time integrating serious bleeding complications into the final outcome. Interestingly,<br />

in order to justify this “new” endpoint, they considered asymptomatic, venographically-confirmed VTE to be clinically<br />

irrelevant and, therefore, could be discounted from the evidence. Needless to say, the arguments against using the<br />

disputed terms symptomatic/clinically relevant and asymptomatic/clinically irrelevant interchangeably are obvious to<br />

every practicing clinician and well summarized in the paper by Bounameaux and Agnelli.<br />

In my opinion, the advantages <strong>of</strong> advancing the discussion about VTE prophylaxis practices as universal recommendation<br />

are many. It certainly highlights the importance <strong>of</strong> thoughtfully considering the risks and benefits <strong>of</strong> pharmacoprophylaxis<br />

in every hospitalized patient and removing the automation from the process. The limitations inherent in the historical<br />

studies used by the authors as basis for their recommendation (whether supporting “symptomatic” or disputing<br />

“asymptomatic” events) call for expanding funding for larger, well conducted prospective studies that address these issues<br />

directly. Further research focused on improving clinically-accepted means <strong>of</strong> enhanced pr<strong>of</strong>iling <strong>of</strong> thrombotic as well as<br />

bleeding risks in pr<strong>of</strong>iling <strong>of</strong> hospitalized patients are also needed.<br />

Conclusions<br />

Clinical guidelines are meant to raise awareness and highlight the public health impact as well as establish a framework<br />

for approaching a specific clinical problem. Clinical judgment and individualized risk assessment remain the most critical<br />

component <strong>of</strong> decision making, and this is particularly true when it comes to VTE prophylaxis.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!